Saturday, May 8, 2010

Too Loose for Too Long...

I've read numerous times that the Fed Funds rate was kept "too low for too long" after the tech bubble burst. Academics, business people, pundits, and even some Fed officials have said this. And, in retrospect, that's to be expected. One can say, once the data is in, that mistakes were made. The question is, is the lesson we can learn from the mistake applicable to future circumstances?

The main issue here is, I think, whether Fed officials (the FOMC) can know when the Fed Funds rate is too low not in retrospect but in the moment. The "Taylor rule," of John Taylor, appears to be a guide. At least, one can know if the Fed Funds rate is low relative to what the Taylor rule would suggest. The problem isn't whether the Fed Funds rate is relatively low compared to the Taylor rule, though. The problem is whether the Fed Funds rate is relatively low (or high) compared to the interest rate set by the market. But that issue brings me to my next point, which is a parallel of the above.

The discussions around "too low for too long" remind me of discussions regarding the existence of an asset price bubble. First, bubbles are always claimed to have existed once the bubble has burst. A quick, but untestable, definition of a bubble is typically stated as unreasonably high asset prices. The question, of course, is what are the fundamentals that drive the price, and has the price become detached from its long-run relationship with the fundamentals? Speaking as a financial economist, bubbles are difficult to test for, and the methods used are numerous.

The parallel between the "too low for too long" and asset bubbles is this: easy to identify in hindsight, but very difficult to identify in the moment. The problem is that it is "in the moment" that is important. Many pundits and other writers have wondered why "we didn't see this coming." Of course, it is well known now that many economists and other people did see "this" coming. The issue was that policy makers were not listening to the nay-sayers. So the first problem is identification, and the second problem is convincing people that the current situation is untenable.

I think that we are again in a period where the Fed has manipulated the Fed Funds rate to a point that is too low, kept it there too long, and that stock prices and, yes, real estate prices are overly inflated. Greece is the harbinger of doom.

Wednesday, May 5, 2010

Before Reform

I guess I've discovered that responding to others' blog posts isn't going to be satisfying anymore. I've read many things about financial reform, some of which makes sense. A lot doesn't. There's a couple of issues that need to be sorted out before "reform" of any kind can take place.

1) Too Big to Fail (TBTF) : Can we please get a working definition on this other than "really big." Boatloads of companies have a high market cap - does that make them too big to fail? If so, what constitutes failure? If MMM loses 10% of its value in, say, a month, will the federal government decide they need to intervene because MMM is too big to fail? Obviously, this is foolishness. I mean for the example to show that size isn't the sole criterion here.

During the crisis, as TBTF was getting tossed around a new nomenclature grew up that tacitly allowed the Fed and TARP and other bailout operations to give money to non-large banks: too Interconnected to fail. Here I'm even more lost. What do I do, count the number of trading partners a bank has? What makes a bank interconnected? I get that Goldman Sachs is the counterparty to a huge amount of trades - so yes they're interconnected. But GS is also in the habit of being really smart. I'm willing to bet that a lot time is spent managing risk, and making sure the trades that expose them to some risks are offset by other trades. So, yes, they have a lot of trading partners - but that does not make them an inherently risky operation.

2) Beyong defining TBTF, or TITF, I would love a definition of "systemic risk." This seems to be related to TITF. Systemic risk, defined in a singularly unhelpful way, would be: "XYZ poses a risk to the system." I guess that means that if a bank fails then this could cause a ripple effect in the banking system and cause a whole bunch of banks to fail. I'm not certain what the failure mechanism is supposed to be: counterparty problems? animal spirits? Or how about regulator panic?

Counterparty risk is faced all the time in trading. That's why, for exchange-traded derivatives, the clearinghouse is the counterparty. The clearinghouse is A) large and B) is pretty much position-neutral on net, since it's the counterparty to longs and shorts on the same products. But, some firms welcome the counterparty risk and use it as a source of (potential) returns. Usually, counterparty risk is fairly singular - one or two of your trading partners might become distressed at any given time, and you'll lose money on some trades. The problem turns up when a whole bunch of your trading partners become distressed at once. But then the question is why did all these firms encounter problems at the same time?

My point here is that a firm isn't an inherent systemic risk just because it has a lot of trading partners. It may, however, be a conduit for risk transference if there is a shock to the economic system. But then, how is that different from any large industrial firm that has numerous suppliers and customers?

I'm afraid we have a long way to go before we can get to some genuine reform. I will have more to say on a variety of issues in future posts - like CDS and standardization, and the institutionalization of TBTF.